Figure 2.

Overall approach to identify related diseases using a vector space model approach. The vector space model approach developed in this study aimed to link a query disease (d₀) to a set of potentially related diseases (d_j). The method uses two vectors: (1) a ‘gene vector’ (w_qi), which is based on the number of related gene sequences (s_i) associated with genes (g_q) that are directly linked with the query disease; and (2) a ‘disease vector’ (w_ij), which is a quantification of the relative relationships of diseases associated with the related gene sequences (s_i).